Kaggle Game Arena · Heads-up Poker

LLM poker leaderboard: AI models at heads-up poker

These are general-purpose language models playing heads-up poker against each other on Kaggle Game Arena, with no game engines to help. This page covers heads-up poker only.

1 record and 1 tracked run or leaderboard for Heads-up Poker only. Nothing on this page is compared with other games. How we verify.

Which AI is best at heads-up poker?

On Kaggle Game Arena: Heads-up Poker, the current #1 is GPT-5.6 Sol: 30 (Mean BB/100), as of Oct 8, 2026 (verified, from the official leaderboard). The ladder only includes language models, not dedicated heads-up poker engines.

Records

Heads-up Poker records
ModelAchievementRun configStatusDate
GPT-5.6 Sol #1 on Kaggle Game Arena Heads-up Poker: 30 Mean BB/100
Last verified: – · Last checked: Oct 18, 2026
How was this run measured?
Model
GPT-5.6 Sol
Game
Heads-up Poker
Project
Kaggle Game Arena
Harness
Kaggle Game Arena harness
Inputs
cards & betting history as text
Tools
none
Assistance
none
Budget
per-model settings not shown
Verification
Verified Read from the official leaderboard, which links replays of every game.
Evidence
Kaggle Game Arena: Heads-up Poker leaderboard (official) (operator)
Dates
evidence dated Oct 8, 2026
Recorded
published Oct 8, 2026 · discovered by AI vs Games Oct 18, 2026
Changes
No change detected since we started tracking.
Permanent link · Submit a correction · Share card (PNG) · JSON
 Kaggle Game Arena harness
cards & betting history as text
 Verified
source
 Oct 8, 2026

Run history

Tracked runs & leaderboards

Kaggle Game Arena: Heads-up Poker RECENT

Heads-up poker tests reasoning with hidden information. Ranked by mean big blinds won per 100 hands (BB/100).

How was this measured? (applies to every row)
Harness
Kaggle Game Arena harness
Inputs
cards & betting history as text
Tools
none
Assistance
none
Budget
per-model settings not shown
Sources
www.kaggle.com
Submit a correction
#ModelMean BB/100StatusAs of
1 GPT-5.6 Sol 30Verified2026-10-08
2 GPT-5.5 25Verified2026-10-08
3 GPT-6 Astra 25Verified2026-10-08
4 Claude Fable 5.1 18Verified2026-10-08
5 Claude Opus 5 10Verified2026-10-08
6 GPT-5.6 Terra 7Verified2026-10-08
7 GPT-5.2 5Verified2026-10-08
8 GPT-5.6 Luna 5Verified2026-10-08
9 Gemini 3.8 Flash 5Verified2026-10-08
10 Gemini 3.6 Flash 3Verified2026-10-08
11 GPT-5.4 1Verified2026-10-08
12 Grok 4.6 0Verified2026-10-08

Last checked: Oct 18, 2026 · auto-update: manual source

This Week in AI vs Games

One short email every Sunday: new records, memorable failures and runs to watch. Free. Every claim is linked to its source.

Preview placeholder. The Beehiiv embed goes here.

Latest issue: Issue #1: The Ender Dragon is down